workflow-node-code

name	workflow-node-code
description	Builds or updates frontend code-node data for deterministic Python or browser tasks.
allowed-tools	["execute_node_skill_script","update_current_graph"]

Workflow Code Node

Use type: code for deterministic processing, API integration, or browser automation.

Node Schema

build_node.data only contains business fields. Do not provide id, type, position, or status; the script generates or fills them. New code node IDs follow the frontend rule: code-<uuid> with a uniqueness suffix only if needed.

Field	Type	Required	Meaning	Notes
title	string	no	Display title	Defaults to 编码节点
description	string	no	Node description
inputs	array	no	Typed function inputs	Defaults to one input string
outputs	array	no	Typed function outputs	Defaults to one code_result object
config	object	no	Code execution settings	Defaults match frontend code node

Each input/output item contains name, type, and optional description.

Config field	Type	Required	Meaning
codeLanguage	string	yes	Must be python
codeCapability	string	yes	python or browser
codeSource	string	yes	sandbox_file, sandbox_snippet, or inline
codeFilePath	string	conditional	Required for sandbox_file
codeEntryFunction	string	yes	Normally main
outputKey	string	no	Primary declared output

Input Variables

Inputs expose exactly two user-facing modes:

1. Custom string: {"field": "separator", "sourceType": "literal", "source": ",", "valueType": "String"}.
2. Upstream reference: {"field": "input", "sourceType": "node", "source": "start.input", "valueType": "String"}.

context is a legacy runtime source type, not a Skill input mode. Do not emit it. The start-node input is referenced as a normal node output, such as start.input.

Before building or changing any node-reference input, execute list_input_sources. For a new node pass the IDs of its intended direct predecessors from the confirmed topology. For an existing node pass node_id. Only use an exact source returned by the script.

For a node mapping, copy the returned source type into both the input type and mapping valueType. For a literal mapping, the input and valueType must both be String.

Output Variables

Supported output types are:

String, Integer, Number, Boolean, Time, Object, Image, Video, Array, Array<String>, Array<Integer>, Array<Number>, Array<Boolean>, Array<Time>, Array<Object>, Array<Image>, and Array<Video>.

Output names must be unique identifiers using letters, numbers, and underscores, and must not start with a number.

The Python main return value is the node output object. Every declared output must be returned under the same top-level key and with a compatible value:

async def main(args: Args) -> Output:
    input_value = args.params["input"]
    return {
        "code_result": {"value": input_value},
        "item_count": 1,
    }

For this example, declare code_result: Object and item_count: Integer. outputKey must be empty or exactly one declared output name. Do not declare outputs that the code does not return.

Default code node generated by build_node:

{
  "id": "code-<uuid>",
  "title": "编码节点",
  "type": "code",
  "description": "在调试沙箱中执行 Python 代码，转换并返回结构化结果。",
  "inputs": [
    {
      "name": "input",
      "type": "String",
      "description": "传入代码的上下文对象"
    }
  ],
  "outputs": [
    {
      "name": "code_result",
      "type": "Object",
      "description": "代码执行结果"
    }
  ],
  "config": {
    "prompt": 
     
     
     
     
     
     
     
     
      
         
         
         
         
      
    
     
     
     
     
     
     
     
     
     
     
     
  

Rules

• Python entry signature is async def main(args: Args) -> Output:.
• Return an object whose keys match declared outputs.
• Never invent a node mapping source. Use list_input_sources.
• inputs and config.inputMappings must have one complete matching entry per input field.
• Keep outputKey aligned with a declared output (code_result by default).
• Browser capability and sandbox files require an available bound sandbox.
• Do not embed secrets in code or node config.
• Prefer LLM nodes for semantic generation and selector nodes for simple branching.

Scripts

list_input_sources

• Path: workflow-node-code/scripts/node.py
• Entry: list_input_sources
• Description: list valid upstream variables for input mappings.
• New-node arguments: {"upstream_node_ids": ["intended-direct-predecessor-id"]}
• Existing-node arguments: {"node_id": "existing-id"}
• Returns: {"sources": [{"source": "node-id.field", "nodeTitle": "...", "field": "...", "type": "...", "description": "..."}], "allowed_source_values": [...]}

build_node

• Path: workflow-node-code/scripts/node.py
• Entry: build_node
• Description: build a complete new code node and generate id and type.
• Arguments: {"data": <Node Schema business fields>, "upstream_node_ids": ["direct-predecessor-id"]}
• Returns: {"node": <complete node>}

update_node

• Path: workflow-node-code/scripts/node.py
• Entry: update_node
• Description: find the current code node by ID and apply changed fields.
• Arguments: {"node_id": "existing-id", "changes": <changed fields>}
• Returns: {"node": <complete updated node>}

Objects merge recursively; arrays replace the entire old array; scalars replace directly; omitted fields stay unchanged. Submit complete arrays when changing inputs, outputs, or config.inputMappings.